Animal model data and clinical studies suggest that exaggerated cardiovascular responses to behavioral stress may promote the development of atherosclerosis and may trigger clinical events. The absence of will designed epidemiologic studies investigating this proposal and the invasive nature of coronary angiography and ethical constraints on the angiographic evaluation of asymptomatic individuals without suspected coronary disease have hampered research on stress-induced cardiovascular responses as a potential risk factor or marker of disease. In this context, ultrasound measurements of carotid atherosclerosis provide an alternative methodology to test the reactivity-atherosclerosis association as these measures can be made reliably and are correlated with atherosclerosis in the coronary arteries. Postmenopausal middle- aged women appear to have significant, albeit subclinical atherosclerosis. We propose to evaluate the association between cardiovascular reactivity to behavioral challenge and carotid artery disease, i.e., plaque index and average wall (intima-media) thickness scores, in an already established sample of women enrolled in an ongoing epidemiological study of the menopause, called the Healthy Women Study. In this study, detailed psychosocial, sociodemographic, biological, and medical use assessments were made in 1983-84 among 541 healthy premenopausal women and were made periodically thereafter as women became peri- and post-menopausal. In 1991, the protocol was altered to include measurements of blood pressure and heart rate changes during two psychological stressors. We now propose to measure carotid atherosclerosis in the first 400 participants who are five years postmenopausal in order to evaluated the reactivity hypothesis, to test if the relationship between stress-induced responses and atherosclerosis vary according to women's exposure to postmenopausal hormone replacement therapy and to other key biological risk factors, especially, waist-hip ratio, resting blood pressure, and fasting insulin and glucose levels. Finally we will examine if the associations between psychosocial characteristics and atherosclerosis are mediated by cardiovascular responses to stress.